基于日志挖掘的查询概念图扩展  

QUERY CONCEPTION GRAPH EXPANSION BASED ON LOG MINING

作　　者：李小琳[1] 陆汝占[1] 

机构地区：[1]上海交通大学计算机科学与工程系,上海200240

出　　处：《计算机应用与软件》2010年第3期1-3,21,共4页Computer Applications and Software

基　　金：国家自然科学基金重大项目(60496326)

摘　　要：提出将概念图引入查询扩展,从概念的层面上进行语义的扩展。使用概念图表示查询可以更准确地表明用户的查询意图,并在此基础上进行语义的扩展,通过这种方法给出的扩展查询更符合用户的查询意图。对用户查询进行基于概念图的查询扩展,并将结果与百度的相似查询进行了比较,证明基于概念图的查询扩展能更准确地把握用户的查询意图。In this paper the authors introduce the conceptual graph into query expansion, which implements semantic expansion from conceptual layer. Using conceptual graph to represent query can express user＇s query intention more exactly, the semantic expansion is performed on this basis, and the expanded queries provided in this way are closer with user＇s intention. In this paper we performed the query expansion for users＇ queries on conceptual graph, and compared the results with Baidu＇s similarity query,it is proved that the conceptual graph - based query expansion can more accurately grasp the query intention of the user.

关 键 词：查询扩展 概念图 日志挖掘 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]